Foot Pain and Disability in Older Women

American Journal of Epidemiology Copyright © 1998 by The Johns Hopkins University School of Hygiene and Public Health All rights reserved

Vol. 148, No. 7

Printed In U.SA.

Foot Pain and Disability in Older Women

Suzanne G. Leveille,1 Jack M. Guralnik,1 Luigi Ferrucci,1-2 Rosemarie Hirsch,1 Eleanor Simonsick,1 and Marc C. Hochberg3

aged; disabled; foot; foot deformities; pain; walking

Foot pain can be acutely disabling for people of any age. However, when foot pain becomes disabling in older adults, the consequences are likely to be more severe, compounded by the effects of comorbidities. Remarkably little is known about the prevalence of foot pain in older persons and the extent to which foot pain may contribute to disability in older adults. Older women are at particular risk for foot pain compared with men because of their higher prevalence of foot deformities and both osteoarthritis and rheumatoid arthritis. Recent national estimates reveal that 60 percent of women aged 65 years and older in the United States have bunions versus 40 percent of older men (1). Epidemiologic research from 40 years ago depicted the greater burden of arthritis experienced by Received for publication October 6, 1997, and accepted for publication February 27, 1998. Abbreviations: ADL, activities of dally living; Cl, confidence interval; IADI, instrumental activities of daily living; OR, odds ratio; WHAS, Women's Health and Aging Study. 1 Epidemiology, Demography, and Biometry Program, National Institute on Aging, Bethesda, MD. 2 Geriatric Department, I Fratidni, National Research Institute (INRCA), Florence. Italy. 3 Departments of Medicine and Epidemiology and Preventive Medicine, University of Maryland, Baltimore, MD. Reprint requests to Dr. Suzanne Leveille, Epidemiology, Demography, and Biometry Program, National Institute on Aging, Gateway Building, Suite 3C-309, 7201 Wisconsin Avenue, Bethesda, MD 20892.

657

women and identified the foot as the fourth most commonly affected site in persons with osteoarthritis (2). One of the very few epidemiologic studies of foot pain and disability reported that older women were more than twice as likely to have foot pain (56 percent) compared with men (24 percent) and that foot pain was associated with decreased walking speed, shorter step length, and a higher prevalence of lowerextremity disability (3). In this study, we sought to determine the prevalence of foot pain and deformity in older disabled women and to examine whether foot pain is associated with disability and functional performance. The Women's Health and Aging Study (WHAS) offered the unique opportunity to study the association between foot problems and lower extremity function and disability, controlling for other conditions affecting the lower limbs, such as confirmed knee and hip osteoarthritis and peripheral arterial disease.

MATERIALS AND METHODS Study population

The WHAS is a population-based, prospective study of the causes and course of moderate-to-severe disability in women aged 65 years and older. An agestratified random sampling of female Medicare bene-

Downloaded from aje.oxfordjournals.org by guest on July 13, 2011

In a study of the relation between foot pain and disability, a cross-sectional analysis was performed using baseline data (1992-1995) from the Women's Hearth and Aging Study, a population-based study of 1,002 disabled women aged 65 years and older living in Baltimore, Maryland. Chronic and severe foot pain, defined as pain lasting 1 month or longer in the previous year, plus pain in the previous month rated severe (7-10 on a scale of 0 to 10), was reported by 14% of the women. Severe foot pain was more common in women who were younger (aged 65-74 years), obese, or had hand or knee osteoarthritis. Walking speed and five repeated chair stands were slower in women with foot pain. After adjustment for age, body mass index, race, education, self-rated hearth, smoking status, comorbidities, and number of other pain sites, severe foot pain was independently associated with increased risk for walking difficulty (adjusted odds ratio = 1.69, 95% confidence interval 1.10-2.59) and disability in activities of daily living (adjusted odds ratio = 1.91, 95% confidence interval 1.21-3.01). These findings suggest that severe foot pain may play a key role in disability in older women. Further studies are warranted to confirm these results longitudinally and to determine whether interventions to alleviate foot pain could reduce or prevent disability in older women. Am J Epidemiol 1998; 148:657-65.

658

Leveille et al.

ficiaries in Baltimore, Maryland, identified 5,516 women. Of these, 4,137 women completed the screening for physical disability, 1,409 were eligible for the study, and 1,002 agreed to participate. Subject screening and baseline evaluations took place in the participants' homes from November 1992 through February 1995. Eligible women reported difficulty in one or more functional tasks in two or more domains of functioning (basic self-care, upper-extremity ability, mobility/exercise tolerance, higher functioning tasks). Details of the study design and methods have been described previously (4, 5). Study participants were representative of the approximately one-third most disabled older women living in the community. Severely cognitively impaired women, those who scored less than 18 on the Mini-Mental State Examination, were excluded (6). Measurements

Am J Epidemiol

Vol. 148, No. 7, 1998


In this study, foot pain was defined using responses to three questions: "During the past year, have you had pain, aching or discomfort in your feet on most days lasting at least one month (yes/no)?"; "during the past month, have you had this or any pain in your feet (yes/no)?"; and "rate the average pain in your feet in the past month using a scale from 0 to 10 where 0 is no pain and 10 is severe or excruciating pain" (Participants were shown a card and asked to select a number from 0 to 10 indicating the severity of pain). Using these three questions, we sought to distinguish women with the most severe and chronic pain from those who had either intermittent severe pain or moderate pain and, in the third group, those with little or no pain. Women who answered yes to the first two questions and rated their pain as 7-10 on the visual analog scale were classified as having chronic and severe foot pain. Women who reported having pain rated as 4 - 6 on the scale were classified as having moderate pain. In addition, if women responded yes to one, but not both, of the first two questions and rated their pain as from 7 to 10, they were classified as having moderate foot pain. Pain reported as 1-3 was classified as mild foot pain. Women who answered no to both of the first two questions or 0 on the visual analog scale were classified as having no foot pain. The latter two groups, no and mild foot pain, were combined for the analysis. Twelve women were missing foot pain information and were excluded from the analysis. Outcome measures of physical performance included the usual-pace walk (meters/second), fast-pace walk (meters/second), both measured on a 4-m course, five chair stand time (seconds), and standing balance tests. The balance score, ranging from 0 to 7, was a hierarchical variable based on performance of three

increasingly difficult standing balance tests for up to 10 seconds: feet positioned side-by-side, semitandem stand, and full tandem stand (7). In mutually exclusive groupings, a score of zero was assigned if the participant was unable to stand unassisted, and a score of 1 was assigned for individuals who could stand unassisted but could not hold the side-by-side stand. A score of 2 was given if the subject performed the side-by-side stand for 1-9 seconds and did no further stands, and a score of 3 was given for those who performed the side-by-side stand for 10 seconds with no further stands. Similarly, there were two scoring levels for each of the other two progressively more difficult stands. A score of 7 indicated that the full tandem stand was held for 10 seconds. Walking difficulty and disability in activities of daily living (ADL) were assessed during the in-home interview. Walking difficulty was a dichotomous variable based on two questions on the amount of difficulty walking across a small room and walking one quarter of a mile (2-3 blocks). Women who reported any difficulty walking across a small room or a lot of difficulty or the inability to walk one quarter of a mile were classified as having walking difficulty. Disability in ADL was based on report of receiving help with one or more of the following: bathing, dressing, eating, transferring from bed to chair, and using the toilet. Subject characteristics, health information, and foot deformities were assessed in an interview, followed by a physical examination by a trained nurse. Body mass index (weight (kg)/height (m2)) was calculated from height and weight measured by a trained interviewer. Prevalence of 17 major chronic conditions was ascertained using complex algorithms for each condition that drew on information from multiple sources, including self-report, medical records, medication use, radiographs, physical examination, physician questionnaires, and hand photographs. Chronic conditions that were assessed included the following: angina; myocardial infarction; congestive heart failure; peripheral arterial disease; stroke; pulmonary disease; diabetes mellitus; cancer (ever); hip fracture; osteoporosis; symptomatic osteoarthritis of the knees, hip, or hands; rheumatoid arthritis; disc disease; spinal stenosis; and Parkinson's disease. Conditions were classified as definite, possible, or not present on the basis of available information. Detailed descriptions of the disease ascertainment methods and the algorithms have been published previously (7). Foot conditions assessed in the nurse examination included bunions, hammer toes, and bilateral ankle edema. A summary pain score was developed from a number of questions on the baseline interview about pain in specific body sites. The purpose of the measure was

Foot Pain and Disability

twofold: to create an adjustment variable to control for confounding by pain at multiple sites and to estimate whether the foot pain was related to a localized or regional musculoskeletal problem or to a more widespread problem with pain, a distinction of rising clinical and epidemiologic importance (8). Pain was assessed in the following: chest, hand, lower back, hip, knee, and calf. Causes of pain in these sites are primarily related to musculoskeletal and circulatory disorders, two leading contributors to disabling pain (9). Pain in any one site was coded as 1; for example, pain in either or both hands was coded as a 1 for hand pain. The pain score was a sum of the number of reported pain sites and ranged from 0 to 6. Data analysis

Am J Epidemiol

Vol. 148, No. 7, 1998

pain was not associated with disability in any of the logistic regression models, so the findings from these models are not presented. RESULTS

Among the 990 women with complete foot pain information, 314 women (32 percent) had foot pain classified as either moderate or chronic and severe. Nearly half of these women, 14 percent of the entire sample, had chronic and severe foot pain. Factors associated with more foot pain included younger age (65-74 years), black race, greater body mass index, poorer self-rated health, and greater number of chronic conditions (table 1). Amount of reported walking did not vary substantially with level of foot pain. Bunions and hammer toes were very common in these older women (70 and 50 percent, respectively). However, there was no evidence that these problems were associated with chronic and severe foot pain (table 2). Medical conditions that were most strongly associated with severe foot pain included bilateral ankle edema and arthritic conditions. Among the women who had definite symptomatic hand or knee osteoarthritis or rheumatoid arthritis, the percent of women with chronic and severe foot pain was nearly twice that of women without these conditions. Modest differences were observed in the percentages of women who had severe foot pain with presence of congestive heart failure and diabetes, but the differences were not statistically significant. To determine whether the relation between foot problems and foot pain was different in women who walked more, we stratified on amount of walking activity. We found no differences in the proportions of women with or those without bunions or hammer toes who reported chronic and severe pain according to how many blocks they walked per week (data not shown). Women who reported pain at multiple sites were more likely to report chronic and severe foot pain, as shown in figure 1. The percentage of women with severe foot pain increased from 3 to 56 percent in those with from 0 through 6 other pain sites reported. Nearly half of the women (49 percent) with severe foot pain reported pain in at least four other sites. In contrast, among women who did not report severe foot pain, only 18 percent had pain in four or more other sites. In tests of physical performance, as foot pain increased, there was a trend toward poorer performance on walking tests and on the timed chair stands, after adjustment for age (table 3). Women with chronic and severe foot pain were more than 10 percent slower in the usual-pace and fast-pace walk tests compared with women with no foot pain {p < 0.05). Balance test


The levels of foot pain were evaluated according to subject characteristics and health indicators. Chisquare tests for between-group differences were used to evaluate these relations. Proportions of women with chronic and severe foot pain, according to the prevalence of chronic conditions, were also compared by using chi-square tests. Adjusted least-squares means for physical performance tests by level of foot pain were obtained by using analysis of variance methods with general linear modeling (GLM procedure, SAS version 6.11, SAS Institute, Inc., Cary, North Carolina). Pairwise t tests were conducted to check for differences in performance test means comparing women who had chronic and severe foot pain with their counterparts who reported mild or no foot pain. Adjusted odds ratios and 95 percent confidence intervals for disability risks related to severe foot pain were obtained from multivariate logistic regression models. In these models, the reference category included all women who did not have chronic and severe foot pain. Each of the variables shown in tables 1 and 2, as well as the summary pain variable, were assessed for their impact on the odds ratios for disability. Risk factors for foot pain or disability that influenced either of the disability outcome odds ratios were included in the final models. A separate variable was included in the models for the 75 women with missing body mass index data. There were only seven women with missing data for any of the other adjustment variables. When individual disease diagnoses, including osteoarthritis of the knee, hip, and hand, were added to models 2 and 3 in table 4, there was no material change in the odds ratios, and thus, diagnoses were not included in the final models. Additional models were run to evaluate the impact of knee and hip pain on the odds ratios. The analysis was also conducted with moderate pain as a separate independent variable and none or mild pain as the reference group, but moderate

659

660

Leveille et al.

TABLE 1. Percent of women in each foot pain category by characteristics and health behaviors, Women's Health and Aging Study, Baltimore, Maryland, 1992-1995 Foot pain levelt Characteristic or behavior

P value}

None of mid (n = 676)

Moderate (n =174)

Chronic, severe (n = 140)

Age group (years) 65-74 75-84 £85

385 305 300

62.1 69.8 74.7

19.5 17.1 15.7

18.4 13.1 9.6

0.005

Race Black NonUack

278 712

63.0 70.4

19.1 17.0

18.0 12.6

0.046

Education (years)